You are on page 1of 50

Claremont Colleges

Scholarship @ Claremont
CMC Senior Teses CMC Student Scholarship
2013
Te Impact of Participatory Notes on the Indian
Rupee Exchange Rate
Rohan Kothari
Claremont McKenna College
Tis Open Access Senior Tesis is brought to you by Scholarship@Claremont. It has been accepted for inclusion in this collection by an authorized
administrator. For more information, please contact scholarship@cuc.claremont.edu.
Recommended Citation
Kothari, Rohan, "Te Impact of Participatory Notes on the Indian Rupee Exchange Rate" (2013). CMC Senior Teses. Paper 654.
htp://scholarship.claremont.edu/cmc_theses/654

CLAREMONT McKENNA COLLEGE



THE EFFECT OF PARTICIPATORY NOTES ON THE INDIAN RUPEE EXCHANGE
RATE












SUBMITTED TO

PROFESSOR THOMAS WILLETT

AND

PROFESSOR LISA MEULBROEK

AND

DEAN GREGORY HESS

BY

ROHAN KOTHARI






FOR

SENIOR THESIS

Spring 2013

April 29, 2013
2

Abstract


Since 1992, India has grown as a global player in the finance world. In spite of its
success, India has not been able to rid itself of potentially harmful practices. One such practice is
the issuing of Participatory Notes (PN) to foreign investors, so that they can anonymously
purchase securities or derivatives listed on the Indian Stock Exchanges. This instrument came
into public view when it accounted for approximately 50% of all foreign portfolio assets in India.
Since then, the laws regarding PNs have evolved to become a more transparent version of the old
rules. Although PN levels are not close to as high as 2007, a rising trend in PNs has been
observed from September to November 2012. Regulation may have helped figure out who the
end PN holder is, but it has not helped mitigate the inherent volatility that some scholars argued
PNs had. This paper follows previous researchers claims to try to resolve the issue using rigorous
econometric analysis. It uses the Vector Auto Regression (VAR) Model to find the coefficient of
change in Participatory Note volume when regressed against the U.S. Dollar Indian Rupee
Nominal Exchange Rate. Although the models results are interpreted, a problem of serial
correlation existed in the model, thus impairing the results.

S
Acknowledgements


I would like to thank Claremont McKenna College for being a home, (really far) away
from home. Before I came here, I had no idea it was possible to learn, while living in North
Quad. Although, I might not leave here with a stellar academic performance, I will leave here
with a much larger, and more diverse, pool of knowledge. I will leave smarter, and much more of
a global citizen than I was before I transferred to CMC. Most importantly, I will leave here
longing to come back.
I would like to thank my teachers and professors for inculcating copious amounts of
knowledge in me, even though Ive never understood how they did it in face of my laissez-faire
attitude. In this regard, I would like to mention Professor Meulbroek, who even though, I was
told, had an aversion to my type, still managed to cultivate a passion for finance in me.
In light of my Senior Thesis, it would be tantamount to treason if I didnt mention the
amazing support I have gotten from the administrative staff at this school. All the way from
Dean Maraa being at my bedside even before I completely realized I was at a hospital,
1
to my
cleaners Lisa and Ilma who always had my back, and Mrs. Morgan, who somehow managed to
keep track of me, and simultaneously, all the other students at school. I would also like to thank
Professor Willett for being my reader under unorthodox circumstances, and for his invaluable
input.
I would like to thank all the friends Ive made at CMC, and the other Claremont Colleges,
for supporting me. Last, but definitely not the least, I would like to thank my parents for all the

1
Sports injury
4
love and support theyve given me throughout my entire life. I wouldnt be writing this paper
without it.

S
Table of Contents



I. Abstract... 2
II. Acknowledgements..... 3
III. List of Abbreviations... 6
IV. Background.. 8
a. 1991 Balance of Payments Crisis ... 8
b.Post-Crisis Reforms... 10
c. Foreign Institutional Investor ... 12
V. Participatory Notes... 16
a. Regulatory Timeline..... 19
VI. Literature Review......... 22
VII. Data and Methodology.. 24
VIII. Analysis ... 28
IX. Discussion ... 33
X. Conclusion .. 35
XI. Works Cited . 37
XII. Appendix A: STATA Input and Output.. 39



6
List of Abbreviations


General

1. AuC Assets Under Custody
2. BoP Balance of Payments
3. BSE Bombay Stock Exchange
4. DFI Development Finance Institutes
5. DTAA Double Tax Avoidance Agreement
6. ECB External Commercial Borrowings
7. FDI - Foreign Direct Investor
8. FERA Foreign Exchange Regulation Act
9. FII Foreign Institutional Investor
10. FVCI Foreign Venture Capital Invesment
11. IMF - International Monetary Fund
12. INR Indian National Rupee
13. KYC Know Your Client
14. MoU Memorandum of Understanding
15. NRI Non-Resident Indian
16. NSE National Stock Exchange
17. ODI Overseas Derivative Instrument
18. PN Participatory Note
19. RBI Reserve Bank of India
20. PIO Person of Indian Origin
21. SEBI Securities and Exchange Board of India
7
22. SEC Securities and Exchange Commision
23. QFI Qualified Foreign Investor
24. USD United States Dollar

Data and Analysis
1. ARCH Autoregressive Conditional Heteroskedasticity
2. GARCH General Autoregressive Conditional Heteroskedasticity
3. M3 Broad Money
4. VAR Vector Autoregression
5. USDXY US Dollar Spot Rate Index
6. AUCNPN Assests under FIIs, not including Participatory Notes


8
Background



India is a unique country in many ways. The structure of Indias economy has evolved in
a unique manner over its independent history. Politically, it has undergone an interesting
metamorphosis. Numerous kingdoms ruled India before the British colonized and unified the
country in the 19
th
century. After achieving independence in 1947, India decided to remain non-
aligned during the Cold War, in its quest to maintain strategic autonomy and economic
independence. This political sentiment had an immense impact on the development of Indias
economy. Indias early leaders instituted a state-controlled economic system, dubbed the
Licence Raj, and it existed from independence until its eventual dismantling at the end of the
Cold War in the early 1990s.
`Licence Raj is translated into English as the rule of licenses. The term refers to the
Soviet-inspired central planning approach that India adopted, following its independence in
1947. Although the private sector was nominally allowed to operate, it could only do so after
facing numerous regulations, permits, and quotas that together served as symbols of Indias
infamous red tape bureaucracy. Effectively, the only significant private industries in India prior
to economic liberalization were firms that were supported by the state through special licenses
and permits.

1991 Balance-of-Payments Crisis
In the 1980s, India ushered in a period of relatively unprecedented economic growth
through a rapid escalation of public expenditure, which was financed by unsustainable
borrowing. By 1985, the Rupee began to depreciate against the dollar, boosting exports. In spite
of this jump in exports, the soaring debt caused an increase in the current account deficit from
9
1.7% of GDP in 1981 to 3.1% of GDP in 1989.
2
In 1990, Indias exports crashed, as a result of
economic slowdowns in Indias biggest export partners. Credit started freezing up as a result of a
downgrade of government debt from investment grade to junk. Indias foreign exchange reserves
had dropped to merely 7 weeks worth of exports by mid 1990, when the Gulf War began. This
war led to a global oil supply shock, and Indian exports began to dry up. On the other hand,
petroleum imports to India increased by more than 2 billion USD.
3
This exacerbation of the
current account deficit spilled over to affect the capital account. Political instability, a credit
crunch, and the current accounts crisis led to a dip in investor confidence. While investors began
to pull their money out, creditors refused to buy short-term debt, and NRIs (Non-resident
indians, or Indian expatriates), the only significant source of foreign portfolio investment at the
time, began to rapidly pull their money out of India
4
. This balance of payments crisis manifested
itself as the foreign exchange reserves fell to three weeks worth of imports. The urgency of this
crisis was evident when the Finance Minister told parliament that the crisis constitutes a serious
threat to the sustainability of growth processes and orderly implementation of our development
programs.
5
In order to avoid default, India used all its gold reserves as collateral to secure a loan
of 1.8 billion dollars from the IMF in January 1991.
6

The Indian government secured the IMF loan during the interim tenure of the Janata Dal
(Secular) government. Elections were scheduled for May 1991, but were delayed till June
because of the assassination of a prominent opposition candidate and former Prime Minister,

2
Inuia, 1991 Countiy Economic Nemoianuum. vol. 1. Woilu Bank, 1991
3
Bery, Suman, Bosworth, Barry P. and Panagariya, Arvind. India Policy Forum 2009/10.
Volume 6: Editors' Summary. Brookings Institute, July 2009
4
Cerra, Valerie, and Sweta Chaman Saxena. "What caused the 1991 currency crisis in India?."
IMF Staff Papers (2002): 395-425.
5
Budget Speech 1991-92. Shri Manmohan Singh, Minister of Finance. 24th July, 1991
http://www.indiabudget.nic.in/bspeech/bs199192.pdf
6
Inuia, 1991 Countiy Economic Nemoianuum. vol. 1. Woilu Bank, 1991
1u
Rajiv Gandhi. In June, the minority coalition government of P.V. Narasimha Rao was elected,
and it pioneered sweeping reforms under its new Finance Minister, the economist and former
RBI governor, Dr. Manmohan Singh. Although some of the improvements were covenants of the
IMF loan, the government indicated its intention to bring about an era of liberalization in all
sectors of the economy, including crucially, in the financial sector.

Post-Crisis Reforms
In order to comply with the IMFs Balance of Payments Manual, India was forced to
undertake many important changes in its fiscal and monetary policy. The High Level Committee
on Balance of Payments (henceforth Rangarajan Committee) and the Committee on Financial
Sector Reforms (henceforth Narasimham Committee), were set up to leverage these changes to
create a strategy for future growth. To rebuild foreign exchange reserves, and begin the
implementation of market reforms, the new government devalued the Rupee by 23% against the
dollar.
7
The Rangarajan committee suggested the introduction of a floating exchange rate, and
current account convertibility. As recommended by the Narasimham committee, the government
also decided to deregulate interest rates, and allowed prices to be determined by the market.
Credit controls were withdrawn, and there was a decrease in the reserve requirements.
8

Entrepreneurship was allowed to thrive when the barriers to entry were lowered, even though
many industries remained under government control. The financial industry was given special
attention because of its central role in promoting investment to drive future growth. The state-
owned banks were consolidated, and Development Finance Institutes that supplied directed

7
Inuia, 1991 Countiy Economic Nemoianuum. vol. 1. Woilu Bank, 1991
8
Bhasin, Niti. Banking and financial markets in India, 1947 to 2007. New Century Publications,
2007.
11
credit were abolished or converted into commercial banks. This paved the way for increased
private competition, and the entry of foreign banks. Foreign investment in India was encouraged
as a result of more conducive laws, lower barriers to entry, and the convertible Rupee. For
example, foreign controlling interest of businesses in India was allowed, for the first time, in
high priority industries and export driven companies.
9
In November 1992, the National Stock
Exchange (NSE) was incorporated in order to implement international best practices in
secondary markets. The NSE provided investors exposure to Capital Markets (Equity) and
Wholesale Debt Markets. The Narasimham committee recommended that the Reserve Bank of
India be made the sole monetary and banking authority, independent of government intervention.
In order to ensure well-rounded reforms, a recently set up regulatory body was given full
statutory power under the Securities and Exchange Board of India (SEBI) Act of 1992. SEBIs
role was modeled on the United States SEC, with a function to protect the interests of investors
in securities and to promote the development of, and to regulate the securities market.
10

Since the effort to introduce free market policies in India, there have been incremental
changes in the structure of the financial sector. Today it is a curious contradiction between
international best practices and archaic rules. Although India is still an emerging market, some
ascribe its step-by-step growth strategy to its resilience against global financial crises. On the
other hand, the continuing lack of transparency in certain peculiar financial instruments has
served as a conduit for political corruption. This paper examines the effect of one such opaque
instrument, Participatory Notes. The next section discusses the prime users of these instruments,
Foreign Institutional Investors.

9
Budget Speech 1991-92. Shri Manmohan Singh, Minister of Finance. 24th July, 1991
http://www.indiabudget.nic.in/bspeech/bs199192.pdf
10
The Securities and Exchange Board of India Act (1992), www.sebi.gov.in/acts/act15ac.pdf
12
Foreign Institutional Investor (FII)
After the collapse of the Licence Raj, India realized that it was time to globalize. In his
1991 interim budget speech, Finance Minister Dr. Manmohan Singh said, we have now reached
a stage in development where we should welcome, rather than fear, foreign investment.
11

Although the official commencement date was 14
th
September 1992, FII inflows began in
January 1993.
12

Today, there are multiple avenues for foreigners to invest in India. These can be
categorized as Direct Investment, Portfolio Investment, Depository Receipts, External
Commercial Borrowing (ECB), and External Assistance.
Direct Investment is defined as equity investments in Indian companies without using the stock
markets. It is divided into investments in listed companies, called Foreign Direct Investment
(FDI), and unlisted companies called Foreign Venture Capital Investment (FVCI). FDI can be
subdivided into two investment routes, the Automatic Route and the Government Route. As one
can infer from their names, the latter requires approval from the Government whereas the former
has pre-existing procedures for investment. Depository Receipts are derivatives that map the
movement of Indian stocks on foreign stock exchanges. Indian companies issue these receipts,
which do not have any end-user restrictions. ECB is method of investment that allows Indian
companies, both private and public, to raise money internationally. This includes bank loans,
trade credit, and other securitized instruments. External Assistance is aid that India receives from
foreign governments. Portfolio Investment is defined as investments in India through the stock
market or mutual funds. There are three classifications of portfolio investors. The first, and

11
Budget Speech 1991-92. Shri Manmohan Singh, Minister of Finance. 24th July, 1991
http://www.indiabudget.nic.in/bspeech/bs199192.pdf
12
Indian Securities Market Volume XIV 2011, www.nseindia.com/content/us/ismr_full2011.pdf
1S
newest is a Qualified Foreign Investor (QFI). QFIs are defined as non-Indians who are residents
of Financial Action Task Force countries, and are required to have a Memorandum of
Understanding (MoU) with SEBI through their country.
13
This is a recently introduced catchall
category for individuals who are not registered through other routes, but can accept Indian tax
obligations.
14
The second type of investor is a Non-Resident Indian (NRI) or a Person of Indian
Origin (PIO). The Indian government officially assigns these self-explanatory terms to people
who possess NRI or PIO identification cards. People from both these categories are allowed to
own a maximum of 5% of an Indian company.
15
They are also allowed to invest in government
securities and corporate debt.
The third, and biggest category by Assets Under Custody (AuC), is the Foreign
Institutional Investor (FII). In order to be considered as an FII, applicants need to have a proven
track record and be overseen by their respective regulatory authority. Although there are
exceptions to that rule, FIIs are generally one of the following institutions: university funds,
endowments, foundations, charitable trusts, pension funds, mutual funds, investment trusts,
insurance firms, reinsurance firms, international or multilateral organizations or one of its
agencies, or a Foreign Governmental Agency. It could also be an asset management company,
investment manager or advisor, bank or institutional portfolio manager, established or
incorporated outside India and proposing to make investments in India on behalf of broad based
funds
16
and its proprietary funds, if any.
17
As of the 30
th
of November 2012, 1752 FIIs are

13
RBI FAQ, http://www.rbi.org.in/scripts/FAQView.aspx?Id=26
14
ibid
15
ibid
16
Broad based fund is defined as one with a minimum of 20 investors, in which the majority
shareholder owns less than 49%
17
Securities And Exchange Board Of India (Foreign Institutional Investors) Regulations, 1995.
www.sebi.gov.in/acts/act07a.html
14
registered with SEBI.
18
Not only do FIIs trade for themselves, but they also trade for their
clients, who are known as sub-accounts. These sub-accounts do not have to be regulated to be
able to register with SEBI. On the contrary, High Net Worth (over $ 50 mil)
19
Individuals can
register as sub-accounts. On November 30
th
2012, SEBI had 6306 sub-accounts registered.
SEBI limits and regulations on FIIs have been amended and relaxed over time. In 1998
FIIs were first allowed access to the debt market. There are also some restrictions on the
investments that FIIs, and their sub-accounts, can make. Foreign Corporates and Foreign
Individuals are allowed to invest up to 5% of a companys paid-up capital. FIIs are allowed to
invest 10% for themselves, and for all the other types of sub-accounts. Including all their sub-
accounts, FIIs are not allowed to hold more than 24% of the total firm capital.
20
This limit can be
lifted up to the statutory limit, on the behest of the Board of Directors of the company in
question. As of today, there are over 290 Indian companies in which FIIs own over 30% of the
paid-up capital.
21



18
http://www.sebi.gov.in/sebiweb/investment/FIITrendsNew.jsp
19
Securities And Exchange Board Of India (Foreign Institutional Investors) Regulations, 1995.
www.sebi.gov.in/acts/act07a.html
20
Indian Securities Market Volume XIV 2011, www.nseindia.com/content/us/ismr_full2011.pdf
21
List of Foreign Investment in Indian Companies. RBI.
http://www.rbi.org.in/scripts/BS_FiiUSer.aspx
1S
22




22
Indian Securities Market Volume XIV 2011, www.nseindia.com/content/us/ismr_full2011.pdf
Figure 1: Net Investments by, and Number of FIIs Registered from 2000-2010
16
Participatory Notes (PN)


In December 2007, Assets under Management of FIIs nearly doubled year-over-year.
Although this would normally be a reason to rejoice, almost 50% of FII flows in the previous
two years were through an opaque financial instrument known as a Participatory Note. This
section introduces and defines this instrument, and weighs its advantages against its
disadvantages.
23

PNs are a type of Offshore Derivative Instrument (ODI), and are unique to a few Asian
countries, including India. The SEBI (FII) Regulations defines an ODI as an instrument which
is issued overseas by a foreign institutional investor against securities held by it that are listed or
proposed to be listed on any recognised stock exchange in India.
24
These ODIs can also be
issued against derivatives, resulting in leveraged exposure to the Indian Stock Market.
25

Although there are other ODIs like equity-linked notes and capped return notes, PNs are the most
common type of ODI.
26

PNs became a conduit for international investment in India since their inception around
1992.
27
In spite of the fact that the official rules on FIIs were released in 1995, these instruments
were attractive for investors who were apprehensive, yet interested to enter the emerging Indian
market. Theses instruments provide easy access to the Indian market. If an investor would like to

23
Mohan, TT Ram. "Neither Dread Nor Encourage Them." Economic and Political Weekly
(2006): 95-99.
24
Securities And Exchange Board Of India (Foreign Institutional Investors) Regulations, 1995.
www.sebi.gov.in/acts/act07a.html
25
Lokeshwarri S.K. The Fading Allure Of P-Notes. Business Daily. January 28, 2011
26
Indian Securities Market Volume XIV 2011, www.nseindia.com/content/us/ismr_full2011.pdf,
pg188
27
Singh, Manmohan. Use of Participatory Notes in Indian Equity Markets and Recent
Regulatory Changes. Vol. 7. International Monetary Fund, 2007.

17
buy securities or derivatives, he sends his order to the FII. This FII then conveys that order to his
broker in India, who actually buys the securities. In return, PNs are issued to the FII, who then
passes them on to its clients.
28
Apart from its ease of use, there are other factors that made a case
for the creation of PNs. These instruments were freely tradable, which provided liquidity in the
market.
29
This liquidity was much required in the recently and partially liberalized Indian
market. Participatory Notes provided a method for Mauritian individuals and companies to take
advantage of the zero Capital Gains Tax accorded by the Double Taxation Avoidance Agreement
(DTAA).
30
Investing through this instrument also significantly lowered the cost of entry into the
Indian markets. Some of these costs include, the money required to establish broker or bank
relations, costs associated with FII registration and subsequent disclosures, and costs associated
with foreign exchange.
31

In the context of the aforementioned cost advantages over FIIs, one can deduce the higher
premium of an individual transaction through PNs versus FIIs.
32
Similarly, some of the other
advantages of PNs are just one side of a double-edged sword.
The first issue with PNs was the capital gains tax arbitrage opportunities that were
created as a result of the DTAA. Onshore investors had to pay taxes up to 40% on short-term
capital gains, compared to 0% in Mauritius. This led to an increase in the number of offshore

28
N., Nikhil Kumar, Implications of Hedge Funds on the Indian Capital Market (August 20,
2007)
29
White Paper on Black Money. Ministry of Finance, Department of Revenue, India. 2012, pg
27
30
Singh, Manmohan. Use of Participatory Notes in Indian Equity Markets and Recent
Regulatory Changes. Vol. 7. International Monetary Fund, 2007.
31
Report of the Expert Group on Encouraging FII Flows and Checking the Vulnerability of
Capital Markets to Speculative Flows, Government of India, Ministry of Finance, Department of
Economic Affairs, New Delhi, November 2005 (also, Lahiri Committee), pg 16
32
N., Nikhil Kumar, Implications of Hedge Funds on the Indian Capital Market (August 20,
2007)
18
firms that registered in the country. Additionally, certified residents of Mauritius also had to pay
minimal corporate tax, and therefore companies began routing their investments in India through
Mauritius.
33
Between April 2000 and March 2011, Mauritius (41.8%) and Singapore (9.17%)
accounted for over half of the cumulative FDI in India.
34
These disproportionate inflows occur
even though 8 years of the data set is cut out, and some arbitrage opportunities were nullified by
2004, through the decrease in the tax rate on Indian onshore Capital Gains and the increase in the
cost of setting up an establishment in Mauritius.
35

An organization that was against PNs from the start was the RBI. In its dissent of the
Expert Group on FIIs, the RBI suggested winding down all PNs and ceasing further issues. They
said that the suspicious and anonymous nature of the cash flows should be reason enough to shut
PNs down. The anonymity could get further magnified once PNs are traded between foreign
investors through a process they call multi-layering.
36
This made it almost impossible to
monitor entry into the Indian Market. In 2004, SEBI passed directives to curb these issues. They
decreed that PNs could only be issued to regulated bodies. However, it was downplayed as a
mere strengthening of Know Your Client (KYC) norms.
37
SEBI rejected RBIs proposal to ban
PNs because they believed the financial community would have seen it as a recessive measure.
38

One problem that the government did not fix during the 2004 amendment was the
plausibility of Round-tripping. Round-tripping is the process by which tax-free illicit or black

33
Singh, Manmohan. Use of Participatory Notes in Indian Equity Markets and Recent
Regulatory Changes. Vol. 7. International Monetary Fund, 2007.
34
White Paper on Black Money. Ministry of Finance, Department of Revenue, India. 2012.
35
Singh, Manmohan. Use of Participatory Notes in Indian Equity Markets and Recent
Regulatory Changes. Vol. 7. International Monetary Fund, 2007.
36
RBIs dissent to EG. White Paper on Black Money. Ministry of Finance, Department of
Revenue, India. 2012. Annex IV
37
Press Release, Securities and Exchange Board of India, January 23
rd
, 2004
38
Vasudevan, A. "A Note on Portfolio Flows into India." Economic and Political Weekly
(2006): 90-92.
19
money leaves India through illegal routes, and is then repatriated to the country with minimal
fees. Since the official definition of regulated bodies includes an individual or entity
39
whose
investment advisory is regulated, PNs provide a perfect path to bring prodigal money back into
the country. The least probable issue, although still plausible, is the use of PNs by terrorists.
High net worth terrorists could purchase PNs through their asset managers, and then gift the
notes as payment, without paying any tax on the transfer. In spite of a longitudinal cross-section
of society raising concerns about Participatory Notes, they are still traded in the market today.
Therefore, to completely understand the situation today, it is important to be cognizant of the
regulatory past.

Regulatory timeline


Participatory notes have caused friction with regulators in the countries that it exists.
Taiwan introduced strong disclosure requirements for PNs in December 1999, but they rescinded
those regulations 6 months later.
40
The rules on PNs in India have been shaped by the regulatory
changes in the past. The idea of an offshore instrument for investment in Indian securities
probably stems from the Double Taxation Avoidance Agreement with Mauritius. However, PNs
first came into existence when the economy opened up in 1992.
41
In conjunction with the
Foreign Exchange Regulation Act (FERA) of 1973, the Government of India guidelines released

39
Singh, Manmohan. Use of Participatory Notes in Indian Equity Markets and Recent
Regulatory Changes. Vol. 7. International Monetary Fund, 2007. Appendix II
40
Report of the Expert Group on Encouraging FII Flows and Checking the Vulnerability of
Capital Markets to Speculative Flows, Government of India, Ministry of Finance, Department of
Economic Affairs, New Delhi, November 2005 (also, Lahiri Committee), pg 45
41
Singh, Manmohan. Use of Participatory Notes in Indian Equity Markets and Recent
Regulatory Changes. Vol. 7. International Monetary Fund, 2007.
2u
on 14
th
September 1992 provided a framework for foreign investment in India.
42
The official
guidelines for PNs came out in 1995 as part of SEBIs rules on FIIs.
Since then there have been a number of amendments to the rules, but only a few of them have
significantly affected PNs.
The first major amendment was probably made in January 2004. The RBI had voiced its
issues with PNs in 2003. It believed that the opacity, and the ease of acquiring, of these
instruments created more risk than benefit for the economic stability of India. Since the end-user
of this instrument was unknown, it could be used not only for fraudulent purposes but also for
malicious purposes, as detailed above. In order to alleviate these qualms, SEBI decided to install
robust Know Your Customer (KYC) norms. They also stipulated that FIIs could only issue them
to other regulated
43
entities. In spite of these changes, the PN volumes continued to rise until
more than half of FII assets were in the form of PNs. These high volumes proved a cause for
concern because PNs were highly attractive instruments even though SEBI controlled the
anonymity of the instrument. The increase in companies routing their investments through
Mauritius alluded to the fact that investors were taking advantage of the tax arbitrage
opportunities presented by PNs. Regulators were also worried simply about the high volumes,
because a mass international exodus from the Indian stock markets could have a major impact on
the market, possibly affecting domestic investor confidence. On October 26
th
, 2007, SEBI issued
a statement in regard to curbing PN flows. These measures were implemented in the amendment
made on 22
nd
May 2008. There were three fundamental changes these measures made. Firstly,
sub-accounts were disallowed from issuing PNs. Secondly, PNs on derivatives had to be

42
Indian Securities Market Volume XIV 2011, www.nseindia.com/content/us/ismr_full2011.pdf
43
Singh, Manmohan. Use of Participatory Notes in Indian Equity Markets and Recent
Regulatory Changes. Vol. 7. International Monetary Fund, 2007.
21
completely wound up by 2009. Thirdly, there was a 40% ceiling imposed on PNs as a percent of
the total AuC. Fourthly, companies that were below this ceiling were allowed to increase their
percentage of PNs by a maximum of 5% each year.
44
These complicated norms created a sense
of unease within foreign investors, and they began to offload more than just their PNs issued on
derivatives. Although this could be explained as the unraveling of derivative included strategies
such as a Protective Put, SEBI decided to rectify the changes made in May. Apart from sub-
accounts not being able to issue PNs, all other changes were revoked.
45
In 2012 the government
decided to cut down the time lag on reporting PN transactions from 6 months to 10 days.
Although PNs are not close to the levels of 2007, they have been increasing in numbers
recently. In light of this fact, it is important to understand the effects of this instrument on the
stability of the Indian Rupee.


44
Securities And Exchange Board Of India (Foreign Institutional Investors) Regulations
amendment 22
nd
May, 2008. www.sebi.gov.in/acts/act07a.html
45
Securities And Exchange Board Of India (Foreign Institutional Investors) Regulations
amendment 10
th
Oct, 2008. www.sebi.gov.in/acts/act07a.html
22
Literature Review


The Narasimham committee of 1991 suggested against allowing investments from NRIs
because of the volatility associated with those cash flows.
46
However, they offer no empirical
evidence confirming this volatility. The Narasimham committee suggested against NRI flows out
of skepticism caused by the mass outflow of NRI investments right before the 1992 Balance of
Payments crisis. Ironically, instruments still exist today, which could have more volatile cash
flows than the NRI flows. Before we understand the effects of those specific instruments, it is
important to understand the overarching effects of FIIs on the Indian stock market. Rai and
Bhanumurthy
47
used monthly data to determine that equity returns are the main determinants of
FII investment. Dhiman
48
confirmed their findings when he concluded that the average daily
returns of the NSE granger caused the daily purchase to sales ratio of FIIs. Mohan
49
pointed out
that FII flows accounted for 69% of all foreign investment in India, as compared to an average of
20% for developing countries. He predicted that FII flows were more volatile than FDI flows,
thus leading to a greater relative volatility in Indian markets as compared to other emerging
markets. However, using an event study methodology in his other study, Mohan
50
found that on
the contrary FII flow volatility did not systematically affect the stock market. He cited the Black
Monday crash of May 2004 when the stock market was resilient to an eight-sigma volatility, in

46
Bhasin, Niti. Banking and financial markets in India, 1947 to 2007. New Century Publications,
2007.
47
Rai, Kulwant, and N. R. Bhanumurthy. "Determinants of foreign institutional investment in
India: The role of return, risk, and inflation." The Developing Economies 42.4 (2004): 479-493.
48
Dhiman, Rahul. Impact Of Foreign Institutional Investor On The Stock Market
International Journal in Research of Finance and Marketing 2.4 (2012):32-46.
49
Mohan, TT Ram. "Stock market fall: Managing volatile flows." Economic and Political
Weekly (2006): 2411-2413.
Su
Mohan, TT Ram. "Neither Dread Nor Encourage Them." Economic and Political Weekly
(2006): 95-99.
2S
spite of FII outflows. Therefore it seems that FII outflows do not create enough instability to
result in a mass sell-of in the Indian market.
The one way that FIIs create weakness in the Indian economy is through the exchange
rate. According to Chandrasekhar
51
large FII inflows lead to an appreciation of the Rupee,
hurting the exports of the country, even though it makes imports cheaper. Conversely, large FII
outflows depreciate the value of the Rupee, increasing exports and making imports more
expensive. In order to mitigate the rise in Rupee, the RBI buys foreign exchange. However, this
affects their ability to control domestic money supply. In order to avoid this problem, before
2006, the RBI would sell government securities equal to the value of foreign exchange
purchased. The Lahiri Report
52
raises concerns about the enhanced volatility of PN flows. It
claims that individuals, who determine their own entry and exit and usually have shorter
investment horizons than investment funds, are able to hold PNs. Further, Singh
53
states that
regulators attribute the cause of the May 2004 and the May/June 2006 FII sell-offs to an initial
sell-off from PN investors.
As a result of its ability to drive other FII flows, PN flows should contribute to a greater
change in the Indian Rupee exchange rate than other FII inflows. Vasudevan
54
states that there
has been no empirical analysis done on the vulnerability of PN cash flows, but unfortunately

S1
Chandrasekhar, C. P. "Courting Risk: Policy Manoeuvres on FII Inflows." Economic and
Political Weekly (2006): 92-95.
S2
Report of the Expert Group on Encouraging FII Flows and Checking the Vulnerability of
Capital Markets to Speculative Flows, Government of India, Ministry of Finance, Department of
Economic Affairs, New Delhi, November 2005 (also, Lahiri Committee)
SS
Singh, Manmohan. Use of Participatory Notes in Indian Equity Markets and Recent
Regulatory Changes. Vol. 7. International Monetary Fund, 2007.
S4
Vasudevan, A. "A Note on Portfolio Flows into India." Economic and Political Weekly
(2006): 90-92.
24
doesnt go on to do any himself. There is very little research about the effect of PNs, specifically,
on the Indian Rupee.
The aim of this paper is to clarify the beliefs about PNs and verify whether it truly creates
more exchange rate volatility than other FII flows.

2S
Data and Methodology


The data used in this empirical study is monthly time-series data from September 2003 to
September 2012. Monthly data was used over quarterly or yearly data because it was the highest
frequency of data that was available. This high frequency of data was preferred because it maps
the regulatory changes and the movement of the exchange rate as closely as possible. The
variables involved in this empirical study are described below. These variables were selected to
try and explain the movement of the exchange rate while controlling for confounding factors.
Individual descriptions and details for the variables are stated below.

United States Dollar/Indian Rupee Exchange Rate (USDINR):
This variable is the nominal exchange rate between the US Dollar and the Indian Rupee.
It was sourced from the Global Financial Database. It is the dependent variable in the regression.
It would be an appropriate measure of currency strength because the performance of the Indian
Rupee is practically always compared to the US Dollar. Similarly, it is also preferred to the real
exchange rate because the nominal exchange rate decides transactions. It was preferred to the
trade-weighted exchange rate because it is easier to isolate all other effects of just one country,
and hence understand the variables that affect the strength of the INR. The change in USDINR
was calculated using the following formula:
!!!"#$%& !
!"#$%&'(
!
!"#$%&'(
!!!
!!
where m is the current month. Closing values at month-end were used.


26
United States Dollar Spot Index (USDXY):
USDINR can be affected by changes in both India and the United States. Therefore, the
USDXY variable is introduced to control against changes in the US Dollar. This variable
represents the index that measures the strength of the US Dollar against a basket of six other
major currencies, not including the Indian Rupee. The change in USDXY was calculated using
the following formula:
!!!"#$% !
!"#$%&'
!
!"#$%&'
!!!
!!
where m is the current month. Closing values at month-end were used.

Consumer Price Index (CPI):
CPI is the total value of the average basket of goods required by a consumer. The percent
change in this number relates to the Inflation for the country. This variable is used to measure
the real effects on the dependent variable, separate from inflation. Data for this was sourced from
the Bureau of Labor Statistics for the United States and from the Labour Bureau for India.
Inflation is given by the formula:
!"#$%&'(" !
!"#
!
!"#
!!!
!!
where m is the current month. Closing values at month-end were used.

Total Value of Participatory Notes (TVPN):
This variable represents the total monetary value of PNs in Rupees Crore (10 million). It
is the most important independent variable in this regression. This variable was the restricting
variable in the regression, because data was only available from September 2003 onwards. This
27
data was derived from the SEBI website. The change in TVPN was calculated using the
following formula:
!!!"#$ !
!"#$%&
!
!"#$%&
!!!
!!
where m is the current month. This change represents the net PN flows each month.

Assets under Custody, Foreign Institutional Investors (AUCFII):
This variable represents the total value of Assets Under FII Custody. The data for this
variable is also sourced from the SEBI website, and has the same restriction as the data for
TVPN.

AUCFII excluding PNs (AUCNPN):
This variable is derived by following formula: !"#$%$ ! !"#$%% !!"#$
Since it is formed from restricting variables, AUCNPN is also restricted. It is used in this specific
way in order to view all other FII investments as separate from PN investments.
The change in AUCNPN was calculated using the following formula:
!!!"#$%$ !
!"#$%&'&
!
!"#$%&'&
!!!
!!
where m is the current month. Similar to TVPN, the change represents the net AUCNPN flows
for each month.

Broad Money (M3):
Broad Money is defined as the sum of currency with the public, demand deposits time
deposits and other deposits with the RBI. An increase in broad money increases the Rupee
28
supply in the Market, and vice versa. The change in Reserve was calculated using the following
formula:
!!!! !
!"#!
!
!"#!
!!!
!!
where m is the current month. The data was sourced from the online RBI database.

RBI Government Securities Turnover (RBI):
Open Market Operations are the market operations conducted by the Reserve Bank of
India by way of sale/ purchase of Government securities to/ from the market with an objective to
adjust the rupee liquidity conditions in the market on a durable basis.
55
A sale of securities
increases the Rupee liquidity in the Market, and vice versa. This variable takes into consideration
the effect of sterilization mentioned in the PN section. The absolute value of turnover was used,
and the data was sourced from the online RBI database.

Note: Broad Money and government securities turnover is reported every week and thus there is
lead or lag in the time of the data by 1-2 days. However, I do not perceive this to be a big issue,
because the securities are sold every Wednesday, and the data is reported every Friday.

55
RBI FAQs - http://www.rbi.org.in/scripts/FAQView.aspx?Id=79#
29


.
dRBI 108 -.3507408 93.57357 -329 480.24
RBI 109 54.81645 102.3259 .35 542.79
dInflation 108 .0000433 .0119457 -.0407474 .0389292
Inflation 109 .0060322 .0092756 -.0389292 .0457516
dchM3 108 -3.85e-07 .0015368 -.0052324 .003895

chM3 109 .0012645 .0010606 -.000453 .0056467
chAUCNPN 109 .0019653 .0075043 -.0229478 .0211312
chTVPN 109 .0016573 .0113841 -.035618 .03202
chUSDXY 109 -.0002915 .0058287 -.0153525 .0171213
chUSDINR 109 .0003673 .0063485 -.0172543 .0177713

Variable Obs Mean Std. Dev. Min Max
Table 1: Summary Statistics - Variables of Interest
Su
Analysis

Note: Appendix A is a STATA supplement to the analysis.
One of the first tests that needs to be conducted with time-series data is a test for
stationary data. This kind of test is also called a unit root test. A unit root occurs when a specific
event in history creates a systematic effect on the regressand. If this effect is not controlled for, a
spurious regression could occur. This could manifest itself as a high R
2
, even if the variables
are uncorrelated.
In order to check for this problem, an Augmented Dickey-Fuller test was performed. This
test is preferred to the basic Dickey-Fuller test because it takes into consideration the lagged
changes in in the variable. Since the data had monthly frequency, a lag of 12 periods was used.
The null hypothesis was that a unit root exists, and the alternative hypothesis stated that it didnt
exist.

Table 2: Augmented Dickey-Fuller Test Results

Number Variable Test Statistic MacKinnon
P-Value
Stationary
1 chUSDINR -3.514 0.0076 Yes
2 chUSDXY -4.596 0.0001 Yes
3 chTVPN -2.703 0.0736 Yes
4 chAUCNPN -3.167 0.0220 Yes
5 chM3 -1.990 0.6067 No
6 RBI -2.666 0.2503 No
7 Inflation -2.360 0.4013 No

S1
The test was significant at the 5% threshold and thus I rejected the null hypothesis for variables
1-4. However, we failed to reject the null for variables 5-7, and the data was not stationary. In
order to try and obtain stationary data for chM3, RBI, and Inflation, I created first differenced
variables using STATA, for each one. The first difference eliminates the effect of the unit root
because the persisting change between the two numbers is eliminated. The Augmented Dickey-
Fuller test was run again on those differenced variables, and the results were as follows:

Table 3: Augmented Dickey-Fuller Test after First Differencing

Number Variable Test Statistic MacKinnon P-
Value
Stationary
1 dchM3 -6.988 0.0000 Yes
2 dRBI -4.480 0.0002 Yes
3 dInflation -5.852 0.0000 Yes

These first-differenced variables were significantly stationary at the 1% critical level. In order to
try and decide how to proceed, one can try and examine the volatility distribution of the
regressand. To meet this, chUSDINR was regressed against the other six dependent variables
according to the following equation:

!!!"#$%& ! !
!
! !
!
! !!!"#$% !!
!
! !!!"#$ !!
!
! !!!"#$%$ !!
!
! !"!!! !!
!
! !"#$ !!
!
! !!"#$%&'(" !!

and then u was plot against Months.
S2


From this graph, it didnt look like error volatility was clustered or conditionally heteroskedastic.
However, this can be evaluated through a post-test. Out of the three vertical separators, the one
on the left is for October 2007, which is when a PN and FII sell-off was triggered when the
Indian Government announced they were going to cap the value of PNs and completely ban the
sale of PNs on derivatives. The middle line represents when the government withdrew the
measures in October 2008, and the one to the right represents the month after the United States
officially
56
came out of its recession. I expected that the exchange rate volatility would have
been larger between the first two bands as a result of the fragile state of the U.S. economy and
foreign inflows into India. Although there is an increase in volatility, the presence of the
USDXY variable probably mitigated the clustering effect. Therefore I decided against using

S6
Accoiuing to the National Buieau of Economic Reseaich
Figuie 2: A plot of the Resiuuals against theii Nonths
SS
Autoregressive Conditional Hetereoskedasticity (ARCH) or General Autoregressive Conditional
Hetereoskedasticity (GARCH) models to begin to fit my data.
The next logical step was to try and use the Vector Auto Regression (VAR) model. The
VAR model compares each variable and its lags to all the other variables and their lags. This
model is a powerful analytic tool just because it allows observation of multiple interactions.
However, before running a VAR, it is important to ensure that there is no co-integration between
any two series over time. Just like non-stationary data, co-integrated time series data could lead
to a spurious regression. To prevent this from occurring, a Johansen test was administered, for
which the lag had to be determined. The varsoc function was used in STATA to generate
Akaikes Information Criterion (AIC) and establish the lag as 1. The Johansen test was
subsequently performed, with the null hypothesis stating that the time series were not co-
integrated. The resulting trace statistic exceeded 5% for all ranks, and therefore we failed to
reject the null hypothesis. The series were not co-integrated, and therefore the unrestricted VAR
model could be implemented using a lag of 1 and 2. STATA created a system of 105 equations,
in which each variable was regressed against 1 lag and 2 lags of all the other variables, and a
constant. The VAR came up with several significant coefficients, some of which stood out. Four
separate variables were correlated with chAUCNPN, which I thought was interesting because it
alludes to the fact that the institutionalized investeor develops his data from multiple sources. I
also found it interesting that as expected, inflation and sometimes lagged inflation correlated
with the change in USDINR and the change in M3. However, both these inferences were
rendered moot by another one. All the lagged variables significantly correlated with their own
variable. This reeked of autocorrelation, and hence I performed a Lagrange Multiplier test of
serial correlation on the VAR. As feared, the statistic turned out to be significant at Lag 2, and I
had to begin again.
S4
Had the VAR worked without errors, there would have been some interesting
relationships to analyze. The most interesting part about a VAR is that one can also observe the
relationship with lagged variables. The first effect that would have been interesting to identify
would be the relationship between the chTVPN and the chUSDINR variable. I would have
expected the coefficient to be positive and significant at the 5 % level. However, it would have
been more interesting to compare the coefficients on chTVPN and chAUCNPN with respect to
chUSDINR. Assuming chAUCNPN would have been significant, this comparison would have
revealed whether PNs had a greater effect on the USD/INR exchange rate than all other kinds of
FII flows. The lagged variable wouldnt have revealed very much justifiable information because
of the one-month frequency.
The lagged relationship between Inflation and chUSDINR, as well as RBI and
chUSDINR would have been interesting to note. Assuming the relationships were significant, it
would have been useful to make sure that both the lagged regressors affected the regressand with
a positive coefficient.

SS
Discussion


One of the limitations that this study had was the monthly frequency of the data.
Exchange rates tend to fluctuate every day. In fact, in Balance of Trade crises, countries like
India have devalued their currencies by over 20% in day. However, the fact that the data was
monthly, helped keep out some of the potential outliers from the daily data.
Probably the biggest flaw in the paper is the serial or autocorrelation, notorious to time-
series data. A simple Breusch-Godfrey statistic brought to light the issue that had existed right
after the first regression was modeled.



After I realized my mistake, I went back and rejected the null hypothesis, and then in order to
account for the serial correlation, I first-differenced the other variables. These variables were
then regressed again, and another Breusch-Godfrey Statistic was found. However, this time the
p-value was even greater than the initial statistic, all of which can be seen in Appendix A. I
believe this effect may have occurred because of three possible reasons: the first could have been
because some of the variables were already in percent changes. The second reason could have
. drop chTVPN[1]
H0: no serial correlation

1 3.636 1 0.0565

lags(p) chi2 df Prob > chi2

Breusch-Godfrey LM test for autocorrelation
. bgodfrey
Table 4: The Breusch-Godfrey LM Statistic for the Initial Linear Regression
S6
been the lower bound on the TVPN and AUCNPN data caused a problem to differencing.
Another possible reason could have been the existence of multicollinearity between the
dependent variables. If it was strong enough, and in a specific direction, it could have thrown off
the standard errors by enough to make the test statistics insignificant.

S7
Conclusion


Although my VAR test failed to provide conclusive data on the experiment, from all the data and
information I have come across, PNs do not seem to have as big as an impact on the exchange
rate as the Indian Media makes it out to have. Initially there was definitely a big reason to be
concerned about PNs. Since the end-user was not known, there were many possible illegal uses
of the instrument, as mentioned in this thesis. Lets look at the question of illegal money re-
entering the country. Although the government ends up loosing tax gains in the short term, it
repatriates money back into India in the long term. In any case the government of India has set
up charity schemes, through which people can begin to convert their illegal money into legal
money by paying a small fee.
From the many non-empirical analyses that I had to go through before I could write this
thesis, I would like to straw man two of the prominent volatility-related arguments that people
make against PNs. The first is the belief that when PN investors buy stocks, they do it as
speculators and not investors. This means they have an approximate investment horizon of less
than a year and bet on the market moving in a certain direction in that short horizon, instead of
expecting the company they invest in to do well in either case. I have two issues with that. For
the sake of argument, lets assume that all PN users are speculators, the speculation can only be
dangerous to the Rupee in the situation of a fear-driven, or irrational dumping of PNs. This is
still no reason to bar a financial instrument. In the worlds largest democracy (by population) I
pity the people who ascribe to cowardly politics instead of welcoming international competition.
The second problem I have with that claim is the fact that foreigners are stereotyped as
speculators. In some situations, firms may want to increase their risk exposure to India, or they
may want to hedge against their currency. Either way for the most part, these decisions are well
S8
thought of investment strategies, in riskier countries with higher costs of Capital. The second and
probably biggest problem I have with PN critics is this belief of Herd Mentality in foreign
investors. There has been some scientific work linking emotionally driven economic decision-
making being linked to information asymmetry, which many foreign investors in India
experience. Although there might be a financial basis for this potential sell-off in the future, it
does not warrant prejudice against the foreign investment community, or one of its vehicles in
India today. In spite of this, in a follow-up study, if it were found that the PNs do have an
adverse effect on the INR, it would be wise for the Indian governments finance ministry to pass
laws better limiting the sale and trade of PNs.




S9
Works Cited

Bery, Suman, Bosworth, Barry P. and Panagariya, Arvind. India Policy Forum 2009/10. Volume
6: Editors' Summary. Brookings Institute, July 2009

Bhasin, Niti. Banking and financial markets in India, 1947 to 2007. New Century Publications,
2007.

Budget Speech 1991-92. Shri Manmohan Singh, Minister of Finance. 24th July, 1991
http://www.indiabudget.nic.in/bspeech/bs199192.pdf

Cerra, Valerie, and Sweta Chaman Saxena. "What caused the 1991 currency crisis in India?."
IMF Staff Papers (2002): 395-425.

Chandrasekhar, C. P. "Courting Risk: Policy Manoeuvres on FII Inflows." Economic and
Political Weekly (2006): 92-95.

Dhiman, Rahul. Impact Of Foreign Institutional Investor On The Stock Market.
International Journal in Research of Finance and Marketing 2.4 (2012):32-46.

Indian Securities Market Volume XIV 2011, www.nseindia.com/content/us/ismr_full2011.pdf

Inuia, 1991 Countiy Economic Nemoianuum. vol. 1. Woilu Bank, 1991

Lokeshwarri S.K. The Fading Allure Of P-Notes. Business Daily. January 28, 2011

Mankiw, Gregory N. "Principles of Macroeconomics, 5th." Ohio: South-Western Cengage
Learning (2009).

Mohan, TT Ram. "Stock market fall: Managing volatile flows." Economic and Political Weekly
(2006): 2411-2413.

Mohan, TT Ram. "Neither Dread Nor Encourage Them." Economic and Political Weekly (2006):
95-99.

Position Paper on External Assistance Received by India, Government of India, Ministry of
Finance, Department of Economic Affairs, New Delhi, March 2008

N., Nikhil Kumar, Implications of Hedge Funds on the Indian Capital Market (August 20, 2007)

Rai, Kulwant, and N. R. Bhanumurthy. "Determinants of foreign institutional investment in
India: The role of return, risk, and inflation." The Developing Economies 42.4 (2004): 479-493.

Report of the Expert Group on Encouraging FII Flows and Checking the Vulnerability of Capital
Markets to Speculative Flows, Government of India, Ministry of Finance, Department of
Economic Affairs, New Delhi, November 2005 (also, Lahiri Committee)

4u
Singh, Manmohan. Use of Participatory Notes in Indian Equity Markets and Recent Regulatory
Changes. Vol. 7. International Monetary Fund, 2007.

Securities And Exchange Board Of India (Foreign Institutional Investors) Regulations, 1995.
www.sebi.gov.in/acts/act07a.html

The Securities and Exchange Board of India Act (1992), www.sebi.gov.in/acts/act15ac.pdf

Vasudevan, A. "A Note on Portfolio Flows into India." Economic and Political Weekly (2006):
90-92.

White Paper on Black Money. Ministry of Finance, Department of Revenue, India. 2012.

Wooldridge, Jeffrey M. Introductory econometrics: A modern approach. South-Western Pub,
2009.






41
Appendix A: STATA OUTPUT and SOME MANIPULATIONS


!"

. dfuller chUSDINR, lags(12)

Augmented Dickey-Fuller test for unit root Number of obs = 95

---------- Interpolated Dickey-Fuller ---------
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value

Z(t) -3.514 -3.517 -2.894 -2.582

MacKinnon approximate p-value for Z(t) = 0.0076


. dfuller chUSDXY, lags(12)

Augmented Dickey-Fuller test for unit root Number of obs = 95

---------- Interpolated Dickey-Fuller ---------
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value

Z(t) -4.596 -3.517 -2.894 -2.582

MacKinnon approximate p-value for Z(t) = 0.0001


. dfuller chTVPN, lags(12)

Augmented Dickey-Fuller test for unit root Number of obs = 95

---------- Interpolated Dickey-Fuller ---------
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value

Z(t) -2.703 -3.517 -2.894 -2.582

MacKinnon approximate p-value for Z(t) = 0.0736


. dfuller chAUCNPN, lags(12)

Augmented Dickey-Fuller test for unit root Number of obs = 95
42

---------- Interpolated Dickey-Fuller ---------
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value

Z(t) -3.167 -3.517 -2.894 -2.582

MacKinnon approximate p-value for Z(t) = 0.0220

. dfuller chM3, lags(12) trend

Augmented Dickey-Fuller test for unit root Number of obs = 95

---------- Interpolated Dickey-Fuller ---------
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value
------------------------------------------------------------------------------
Z(t) -1.990 -4.051 -3.455 -3.153
------------------------------------------------------------------------------
MacKinnon approximate p-value for Z(t) = 0.6067


. dfuller RBI, lags(12) trend

Augmented Dickey-Fuller test for unit root Number of obs = 95

---------- Interpolated Dickey-Fuller ---------
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value
------------------------------------------------------------------------------
Z(t) -2.666 -4.051 -3.455 -3.153
------------------------------------------------------------------------------
MacKinnon approximate p-value for Z(t) = 0.2503


. dfuller Inflation, lags(12) trend

Augmented Dickey-Fuller test for unit root Number of obs = 95

---------- Interpolated Dickey-Fuller ---------
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value
------------------------------------------------------------------------------
Z(t) -2.360 -4.051 -3.455 -3.153
------------------------------------------------------------------------------
MacKinnon approximate p-value for Z(t) = 0.4013

4S





. dfuller dchM3, lags(12)

Augmented Dickey-Fuller test for unit root Number of obs = 94

---------- Interpolated Dickey-Fuller ---------
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value
------------------------------------------------------------------------------
Z(t) -6.988 -3.518 -2.895 -2.582
------------------------------------------------------------------------------
MacKinnon approximate p-value for Z(t) = 0.0000

. dfuller dRBI, lags(12)

Augmented Dickey-Fuller test for unit root Number of obs = 94

---------- Interpolated Dickey-Fuller ---------
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value
------------------------------------------------------------------------------
Z(t) -4.480 -3.518 -2.895 -2.582
------------------------------------------------------------------------------
MacKinnon approximate p-value for Z(t) = 0.0002

. dfuller dInflation, lags(12)

Augmented Dickey-Fuller test for unit root Number of obs = 94

---------- Interpolated Dickey-Fuller ---------
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value
------------------------------------------------------------------------------
Z(t) -5.852 -3.518 -2.895 -2.582
------------------------------------------------------------------------------
MacKinnon approximate p-value for Z(t) = 0.0000







44
$"



. pieuict ies1, i
(S missing values geneiateu)

.

















_cons .0011491 .0004691 2.45 0.016 .0002185 .0020796
dRBI 3.98e-07 4.93e-06 0.08 0.936 -9.39e-06 .0000102
dInflation -.0549006 .0388463 -1.41 0.161 -.1319612 .0221599
dchM3 -.1968207 .3013988 -0.65 0.515 -.7947148 .4010733
chAUCNPN -.2260905 .0698298 -3.24 0.002 -.364614 -.087567
chTVPN -.1123166 .0447407 -2.51 0.014 -.2010701 -.0235631
chUSDXY .4747286 .0832284 5.70 0.000 .3096259 .6398313

chUSDINR Coef. Std. Err. t P>|t| [95% Conf. Interval]

Total .004352597 107 .000040678 Root MSE = .0047
Adj R-squared = 0.4559
Residual .002235628 101 .000022135 R-squared = 0.4864
Model .002116969 6 .000352828 Prob > F = 0.0000
F( 6, 101) = 15.94
Source SS df MS Number of obs = 108
. reg chUSDINR chUSDXY chTVPN chAUCNPN dchM3 dInflation dRBI
4S
%"&
giaph twoway line ies1 Nonth



'"





.
Exogenous: chUSDXY chTVPN chAUCNPN dchM3 dRBI dInflation _cons
Endogenous: chUSDINR

4 418.586 2.5433 1 0.111 .000023 -7.83819 -7.72487 -7.55849
3 417.314 .16348 1 0.686 .000023 -7.83296 -7.72995 -7.5787
2 417.232 1.6845 1 0.194 .000023 -7.85062 -7.75791 -7.62178
1 416.39 .45564 1 0.500 .000023 -7.85366 -7.77125 -7.65024
0 416.162 .000022* -7.86851* -7.7964* -7.69052*

lag LL LR df p FPE AIC HQIC SBIC

Sample: Jan 2004 - Aug 2012 Number of obs = 104
Selection-order criteria
. varsoc chUSDINR, exog(chUSDXY chTVPN chAUCNPN dchM3 dRBI dInflation)
46
("





)"



7 56 1805.3922 0.36180
6 55 1781.8147 0.44548 47.1551 3.76
5 52 1750.8581 0.56070 109.0683 15.41
4 47 1707.673 0.63082 195.4384 29.68
3 40 1655.3586 0.66074 300.0672 47.21
2 31 1598.607 0.73330 413.5705 68.52
1 20 1529.2217 0.80987 552.3411 94.15
0 7 1442.068 . 726.6484 124.24
rank parms LL eigenvalue statistic value
maximum trace critical
5%

Sample: Dec 2003 - Aug 2012 Lags = 1
Trend: constant Number of obs = 105
Johansen tests for cointegration
. vecrank chUSDINR chUSDXY chTVPN chAUCNPN dchM3 dRBI dInflation, lag(1)

_cons .0001856 .0006273 0.30 0.767 -.0010438 .001415

L2. -6.84e-06 6.43e-06 -1.06 0.287 -.0000195 5.77e-06
L1. -1.52e-06 6.69e-06 -0.23 0.820 -.0000146 .0000116
dRBI

L2. .0868708 .0589421 1.47 0.141 -.0286535 .2023951
L1. .1772865 .0546875 3.24 0.001 .0701011 .284472
dInflation

L2. .2766459 .4274654 0.65 0.518 -.5611709 1.114463
L1. .6210479 .4468315 1.39 0.165 -.2547257 1.496821
dchM3

L2. .119262 .0911398 1.31 0.191 -.0593687 .2978928
L1. .0763586 .0938955 0.81 0.416 -.1076733 .2603904
chAUCNPN

L2. -.074831 .0587907 -1.27 0.203 -.1900586 .0403966
L1. -.0584354 .061371 -0.95 0.341 -.1787204 .0618495
chTVPN

L2. .1658357 .1191602 1.39 0.164 -.0677141 .3993855
L1. -.0672624 .1232086 -0.55 0.585 -.3087469 .1742221
chUSDXY

L2. -.1783786 .1387755 -1.29 0.199 -.4503737 .0936165
L1. .2460982 .1312475 1.88 0.061 -.0111422 .5033386
chUSDINR
chUSDINR

Coef. Std. Err. z P>|z| [95% Conf. Interval]


dRBI 15 86.062 0.2591 37.06522 0.0007
dInflation 15 .010497 0.3389 54.34967 0.0000
dchM3 15 .001276 0.4019 71.22096 0.0000
chAUCNPN 15 .007423 0.1434 17.74913 0.2184
chTVPN 15 .010849 0.2323 32.07855 0.0039
chUSDXY 15 .006102 0.0680 7.730514 0.9029
chUSDINR 15 .006171 0.2023 26.88937 0.0199

Equation Parms RMSE R-sq chi2 P>chi2
Det(Sigma_ml) = 1.36e-24 SBIC = -30.47234
FPE = 9.97e-24 HQIC = -32.04133
Log likelihood = 1859.864 AIC = -33.11065
Sample: Dec 2003 - Sep 2012 No. of obs = 106
Vector autoregression
. var chUSDINR chUSDXY chTVPN chAUCNPN dchM3 dInflation dRBI, lag(1 2)
47









L2. -.1783786 .1387755 -1.29 0.199 -.4503737 .0936165
L1. .2460982 .1312475 1.88 0.061 -.0111422 .5033386
chUSDINR
chUSDINR

Coef. Std. Err. z P>|z| [95% Conf. Interval]
L2. .0868708 .0589421 1.47 0.141 -.0286535 .2023951
L1. .1772865 .0546875 3.24 0.001 .0701011 .284472
dInflation
chUSDINR
chUSDXY

L2. -.052197 .0581317 -0.90 0.369 -.1661331 .0617391
L1. -.1132866 .0606831 -1.87 0.062 -.2322233 .00565
chTVPN
L2. -.0200228 .2439804 -0.08 0.935 -.4982156 .45817
L1. -.7192643 .2307454 -3.12 0.002 -1.171517 -.2670116
chUSDINR
chTVPN


_cons .0019712 .0011028 1.79 0.074 -.0001902 .0041326
L1. -.175022 .1578714 -1.11 0.268 -.4844441 .1344002
chUSDINR
chAUCNPN


L2. .3280637 .1669265 1.97 0.049 .0008939 .6552336
L1. -.0353152 .1482018 -0.24 0.812 -.3257854 .2551549
chUSDXY

L2. -.2620619 .1433322 -1.83 0.067 -.5429878 .0188639
L2. .0964713 .0707165 1.36 0.173 -.0421305 .2350731
L1. .1743466 .0738202 2.36 0.018 .0296617 .3190316
chTVPN

_cons .0016634 .0007545 2.20 0.027 .0001846 .0031422
chUSDINR
dchM3


L2. -.0293989 .0121867 -2.41 0.016 -.0532844 -.0055133
L1. -.0382377 .0113071 -3.38 0.001 -.0603992 -.0160763
dInflation

L2. -.260749 .088382 -2.95 0.003 -.4339745 -.0875236
L1. -.6366976 .092386 -6.89 0.000 -.8177709 -.4556243
dchM3
chUSDINR
dInflation


L2. -.2799954 .1002627 -2.79 0.005 -.4765067 -.083484
L1. -.5587198 .0930255 -6.01 0.000 -.7410464 -.3763932
dInflation
48




*"


+"











L2. -3396.753 1935.392 -1.76 0.079 -7190.052 396.5455
L1. 4473.99 1830.405 2.44 0.015 886.4626 8061.517
chUSDINR
dRBI


L2. -.1681707 .0897174 -1.87 0.061 -.3440136 .0076722
L1. -.2825879 .093232 -3.03 0.002 -.4653192 -.0998565
dRBI
H0: no autocorrelation at lag order

2 64.5057 49 0.06785
1 58.6423 49 0.16281

lag chi2 df Prob > chi2

Lagrange-multiplier test
. varlmar
. drop chTVPN[1]
H0: no serial correlation

1 3.636 1 0.0565

lags(p) chi2 df Prob > chi2

Breusch-Godfrey LM test for autocorrelation
. bgodfrey
49
,"


H0: no serial correlation

1 13.004 1 0.0003

lags(p) chi2 df Prob > chi2

Breusch-Godfrey LM test for autocorrelation
. bgodfrey

d2RBI -2.44e-06 3.95e-06 -0.62 0.538 -.0000103 5.39e-06
d2Inf -.0662863 .0291876 -2.27 0.025 -.1241866 -.008386
d2M3 -.1614567 .2231376 -0.72 0.471 -.6041018 .2811884
dchAUCNPN -.235586 .0591257 -3.98 0.000 -.3528755 -.1182965
dchTVPN -.0565114 .0399012 -1.42 0.160 -.1356646 .0226419
dchUSDXY .4866393 .0715082 6.81 0.000 .3447863 .6284922

dchUSDINR Coef. Std. Err. t P>|t| [95% Conf. Interval]

Total .00701796 107 .000065588 Root MSE = .00591
Adj R-squared = 0.4684
Residual .003521833 101 .00003487 R-squared = 0.4982
Model .003496127 6 .000582688 Prob > F = 0.0000
F( 6, 101) = 16.71
Source SS df MS Number of obs = 107
. reg dchUSDINR dchUSDXY dchTVPN dchAUCNPN d2M3 d2Inf d2RBI, noc